Development based on software architecture is one of the most effective solutions to improve software qual-ity and productivity, and minimize the difficulties of developing large and complex systems. Formal models and...Development based on software architecture is one of the most effective solutions to improve software qual-ity and productivity, and minimize the difficulties of developing large and complex systems. Formal models and speci-fications of software architecture are one of the key research areas of software architecture. In this paper, the exten-sion of XYZ/E Language based on set theory, called XYZ/E+ Language, is proposed, and a method of formalizingsoftware architecture whit it is formulated. The formal method is distinguished by its ability to represent not onlystatic properties but also dynamic behaviors, and its high-level abstraction and powerful expressiveness. Finally,through an example of formalizing software architecture of a specific system with XYZ/E+ Language, its powerfulmodeling ability and good practicality are demonstrated.展开更多
适应性是人工免疫系统(AIS,Artificial I mmune System)的重要特性之一。在AIS软件开发应用中,数据源的进化和学习算法的进化是两个有复杂关联的适应性问题。为此我们扩展并改进了已有的AIS构架,提出一个新的适应性软件构架。该构架以...适应性是人工免疫系统(AIS,Artificial I mmune System)的重要特性之一。在AIS软件开发应用中,数据源的进化和学习算法的进化是两个有复杂关联的适应性问题。为此我们扩展并改进了已有的AIS构架,提出一个新的适应性软件构架。该构架以基因计算为中心,扩展了元基因来适应数据源的进化,并设计了可接入学习算法构件和算法验证机制来解决算法进化的适应性问题。在该构架支持下,数据源的进化独立于学习算法的设计,同时使学习算法能适用于多种数据源且能独立进化。该构架可简化AIS软件的复杂性,可提高AIS开发应用的效率,也有助于实现将来的自适应的免疫计算。展开更多
文摘Development based on software architecture is one of the most effective solutions to improve software qual-ity and productivity, and minimize the difficulties of developing large and complex systems. Formal models and speci-fications of software architecture are one of the key research areas of software architecture. In this paper, the exten-sion of XYZ/E Language based on set theory, called XYZ/E+ Language, is proposed, and a method of formalizingsoftware architecture whit it is formulated. The formal method is distinguished by its ability to represent not onlystatic properties but also dynamic behaviors, and its high-level abstraction and powerful expressiveness. Finally,through an example of formalizing software architecture of a specific system with XYZ/E+ Language, its powerfulmodeling ability and good practicality are demonstrated.
文摘适应性是人工免疫系统(AIS,Artificial I mmune System)的重要特性之一。在AIS软件开发应用中,数据源的进化和学习算法的进化是两个有复杂关联的适应性问题。为此我们扩展并改进了已有的AIS构架,提出一个新的适应性软件构架。该构架以基因计算为中心,扩展了元基因来适应数据源的进化,并设计了可接入学习算法构件和算法验证机制来解决算法进化的适应性问题。在该构架支持下,数据源的进化独立于学习算法的设计,同时使学习算法能适用于多种数据源且能独立进化。该构架可简化AIS软件的复杂性,可提高AIS开发应用的效率,也有助于实现将来的自适应的免疫计算。
